- Email: [email protected]
Relationship Between Interdialytic Weight Gain and Nutritional Markers in Younger and Older Hemodialysis Patients
Relationship Between Interdialytic Weight Gain and Nutritional Markers in Younger and Older Hemodialysis Patients Su-Chen Yang, BS,* Chih-Kang Chiang, MD, PhD,† Shih-Ping Hsu, MD,†‡* and Kuan-Yu Hung, MD, PhD‡ Objective: Our objective was to investigate the correlations of interdialytic weight gain (IDWG) with the Malnutrition Inflammation Score (MIS) and other nutritional markers, especially when age difference is considered. Design: This was a cross-sectional study. Setting: The setting was an outpatient hemodialysis (HD) center in a community hospital. Patients: Excluding those with obvious inflammatory diseases, hospitalizations, and major surgery within the preceding 3 months, we enrolled all anuric patients who had undergone maintenance hemodialysis three times a week for .1 year. In total, 84 women and 80 men were enrolled. Their age (mean 6 SD) was 57.9 6 13.0 years. Main Outcome Measures: The conditions of food intake and nutritional status were evaluated with MIS items. Concerning IDWG, the average of 12 sessions within 4 weeks was used. The relative IDWG (RIDWG) was calculated as IDWG divided by the respective dry weight. Other laboratory data were obtained from routine monthly sampling. Results: Whereas IDWG and RIDWG had no significant correlation with any of the MIS items in younger patients (,65 years old; n 5 106), RIDWG had a positive correlation with the severity of insufficient food intake, gastrointestinal upset, functional incapacity, and wasting of muscle and subcutaneous fat in older patients ($65 years old; n 5 58). On the other hand, IDWG and RIDWG had no significant correlation with serum albumin level in younger and older HD patients, respectively. Conclusions: Our findings suggest that in older HD patients, the greater that the RIDWG is, the poorer the nutritional status will be. However, there is still controversy regarding IDWG and RIDWG as nutritional markers in HD patients. Ó 2008 by the National Kidney Foundation, Inc.
P
ATIENTS UNDERGOING maintenance hemodialysis (HD) are always advised to control their fluid intake strictly in order to prevent long-term adverse cardiovascular outcomes and intradialytic hypotension because of large ultrafiltration.1–5 However, out of fear that inadequate food intake will worsen nutritional status, some patients always have greater interdialytic weight *Department of Nursing, Far Eastern Memorial Hospital, Taipei, Taiwan. †Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan. ‡Department of Internal Medicine, Far Eastern Memorial Hospital, Taipei, Taiwan. S.-C.Y. and C.-K.C. contributed equally to this article. Address reprint requests to Shih-Ping Hsu, MD, Department of Internal Medicine, Far Eastern Memorial Hospital, No. 21, Sec. 2, Nan-Ya South Road, Pan-Chiao, Taipei 244, Taiwan.
E-mail: [email protected] 2008 by the National Kidney Foundation, Inc. 1051-2276/08/1802-0006$34.00/0 doi:10.1053/j.jrn.2007.11.012
Ó
210
gain (IDWG) than advised. Conversely, other patients with less of an IDWG actually have inadequate food intake and poor nutritional status. Therefore, it is clinically interesting and important to determine whether IDWG reflects caloric and protein intake and correlates with other nutritional markers in HD patients. In the literature, some studies showed that IDWG is positively correlated with serum albumin levels in HD patients.6–8 Furthermore, it was proposed that HD patients with a large IDWG have better survival outcomes.6 Yet the relationship between IDWG and subjective global assessment (SGA)9 parameters has not been well-discussed. Thus, in this cross-sectional study, we attempted to investigate correlations of IDWG with SGA parameters and other nutritional markers in chronic HD patients, taking age differences into particular account. For comprehensive coverage, the Malnutrition Inflammation Score (MIS),10 an extended and modified form of SGA, was adopted. Journal of Renal Nutrition, Vol 18, No 2 (March), 2008: pp 210–222
IDWG AND NUTRITION IN HD
Subjects and Methods Patients To minimize the confounding effects of residual renal function on IDWG and nutritional markers,11 we enrolled only anuric (,50 mL/ day) HD patients who had undergone HD three times a week for .12 months at the outpatient HD unit of the Far Eastern Memorial Hospital (Taipei, Taiwan). Considering known confounding factors for nutritional status, we excluded patients with the following conditions or events within the preceding 3 months: obvious infections or inflammatory diseases, hospitalizations, and major surgery. In total, 164 patients were enrolled. This clinical study adhered to the Declaration of Helsinki, and informed consent was obtained from patients. Regarding the demographics and etiology of end-stage renal disease (ESRD), we obtained enrollment data from the medical charts and computerized database at the HD unit. Laboratory Parameters All biochemical analyses were performed with a Hitachi 747 Automatic Analyzer (Hitachi, Tokyo, Japan). The results of routine monthly midweek, nonfasting, predialysis blood samples were used for analysis. Postdialysis blood urea nitrogen (BUN) levels of the same dialysis session were measured to calculate the delivered dose of dialysis (Kt/V) by the method of Daugirdas.12 The predialysis BUN level of the next dialysis was measured to calculate the normalized protein catabolic rate (nPCR, normalized to body weight derived from the urea distribution space [Vurea/ 0.58]). Serum calcium levels were corrected for serum albumin, according to the following formula: collected serum Ca (mmol/L) 5 serum calcium (mmol/L) 1 0.02 3 [40 – serum albumin (g/L)]. Dialysis Therapy All patients had undergone HD three times a week with bicarbonate-based dialysate, volumetric ultrafiltration control, and single-use, cellulose-based membranes. The dry weight of each patient was determined on a trial-and-error basis and reevaluated frequently. Each patient was visited by our nephrologists at each HD session. The session time lasted 4.0 6 0.3 hours (range, 3.0 to 4.5 hours). The prescriptions of recombi-
211
nant Human Erythropoietin and iron supplement were based on a standardized algorithm, with the target Hematocrit at 30% to 36%.
Interdialytic Weight Gain and Relative Interdialytic Weight Gain With regard to IDWG, the average of all 12 sessions within the previous 4 weeks before enrollment was used for analysis. The relative IDWG (RIDWG) was calculated as the average IDWG divided by the respective dry weight. Malnutrition Inflammation Score (Subjective Global Assessment) Proposed by Kalantar-Zadeh et al10 as a comprehensive, risk-stratifying scoring system for morbidity and mortality in maintenance HD patients, the Malnutrition Inflammation Score (MIS) integrates patients’ nutritional history, physical examination findings, body mass index, and laboratory parameters. Concerning patients’ nutritional history, the following five components, adopted from the original subjective global assessment (SGA),9 were taken into account: overall change in end-dialysis dry weight in the past 5 months, dietary intake, gastrointestinal (GI) symptoms, impairment of nutritionally related functional capacity, and comorbidity, including number of years on dialysis. Based on the SGA criteria, physical examinations involved two observations, ie, decreased fat stores or loss of subcutaneous fat, and signs of muscle-wasting. Body mass index (BMI) was calculated as end-dialysis body weight in kilograms divided by the square of body height in meters. Serum albumin levels and total iron-binding capacity (TIBC) were selected as the pertinent and representative laboratory parameters. Thus, the MIS has a total of 10 components, each with four levels of severity, from 0 (normal) to 3 (severely abnormal). The sum of all 10 MIS components ranges from 0 (normal) to 30 (severely malnourished). A higher score reflects a more severe degree of malnutrition and inflammation. Through the courtesy of K. Kalantar-Zadeh, detailed definitions of each component of MIS have been supplied as an Appendix. With reference to the original scoring method proposed by Kalantar-Zadeh et al,10 we assigned a trained physician’s assistant to score each patient within 5 to 15 minutes before anthropometric
212
YANG ET AL
measurements were performed. To evaluate the degree of reproducibility, the same physician’s assistant repeated the MIS assessment after 1 week on a subset of 20 patients without reference to their first MIS evaluation. Compared with the findings in the previous report, the correlation coefficient (r) between the two MIS assessments was 0.90, denoting a good degree of reproducibility.
Anthropometric Evaluation Another trained physician’s assistant performed anthropometric measurements 5 to 20 minutes after the termination of an HD session. These measurements included triceps skinfold thickness (TSF), mid-arm circumference (MAC), and derived mid-arm muscle circumference (MAMC). All anthropometric measurements were performed by the trained physician’s assistant three times in rapid succession on the nonaccess-containing arm of each dialysis patient, and the average of the three measurements was regarded as the final result for analysis.
main causes of ESRD. In these patients, the IDWG was 2.9 6 1.1 kg, and the RIDWG was 5.3% 6 1.7%. In almost all components of the MIS, most of the patients scored 0, ie, normal. The exception was the component of ‘‘comorbidity’’ in which the score was at least 1, because we only enrolled patients who had undergone HD for .1 year. The sum of the MIS ranged from 1 to 22, with a mean of 6.4 6 3.9.
Results
Relationship of IDWG and RIDWG With Severity of Individual Components of MIS and Other Relevant Parameters At first, the relationship of IDWG and RIDWG with the severity of each MIS component and other relevant parameters was analyzed in the group of all enrolled patients (n 5 164). As shown in Table 2, IDWG was negatively correlated with change in dry weight and serum TIBC levels, whereas RIDWG was positively correlated with the severity of poor intake, GI symptoms, functional incapacity, and wasting of muscle and subcutaneous fat. With the sum of the MIS, IDWG had a negative correlation, whereas RIDWG paradoxically had a positive correlation. As shown in Table 3, whereas IDWG was negatively correlated with dialysis clearance (Kt/V), RIDWG was not significantly correlated with Kt/V. Regarding the traditional nutritional markers, MAMC, serum levels of albumin, and TIBC were positively correlated with IDWG, but were not significantly correlated with RIDWG. On the other hand, both IDWG and RIDWG were positively correlated with serum phosphate levels, and negatively correlated with serum sodium levels. The correlations of IDWG with serum albumin, MAMC, and BMI in all patients are shown in Figure 1. Those of RIDWG are shown in Figure 2.
Characteristics of Patients and Laboratory Data The demographic characteristics, dialysis parameters, and biochemical data of patients are presented in Table 1. Of 164 patients, 84 (51.2%) were women, and 80 (48.8%) were men. Their age was 58.7 6 12.9 years (range, 24.0 to 85.8 years). The duration of HD therapy ranged from 12.2 to 269.6 months. Diabetic nephropathy (34.8%) and chronic glomerulonephritis (40.9%) were the
Age Differences in the Correlation of IDWG and RIDWG With MIS Components and Other Relevant Parameters To determine age differences in the correlation of IDWG and RIDWG with MIS components and other relevant parameters, we arbitrarily divided the enrolled patients (n 5 164) into three age groups: #45 years (n 5 22), ie, the young group; between 45 and 65 years (n 5 84), ie, the
Statistical Analysis Unless otherwise stated, continuous variables are presented as mean values 6 standard deviation (SD), and categorical variables as a number for each item. For comparing differences between age groups, Student’s t-test and the chi-square method were used as indicated. For evaluating correlations between IDWG and RIDWG and the severity of individual components in the MIS, the Pearson correlation method was used. A two-tailed P , .05 was considered statistically significant, and a P value of between .05 and .10 was considered marginally significant. All computations were performed with SPSS 10.0 for Windows (SPSS, Inc., Chicago, IL).
213
IDWG AND NUTRITION IN HD
Table 1. Demographic Characteristics and Biochemical Data of All Patients, Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients Variables Demographics Age (y) (range, y) Sex (male/female) Duration on HD (mo) (range, mo) Dry weight (kg) (range, kg) Body mass index (kg/m2) IDWG (kg) RIDWG (% of dry weight) Etiology of ESRD Diabetic nephropathy Chronic glomerulonephritis Others Undetermined Delivered dose of dialysis Daugirdas12 Kt/V URR (%) Anthropometry MAC (cm) TSF (cm) MAMC (cm) Nutrition Albumin (g/L) TIBC (g/L) nPCR (g/kg/day) Laboratory data Hemoglobin (g/dL) Predialysis BUN (mmol/L) Creatinine (mmol/L) Uric acid (mmol/L) ALT (IU/L) cCa (mmol/L) P (mmol/L) cCa 3 P iPTH (ng/L) (range) Na (meq/L) K (meq/L) Total cholesterol (mmol/L) Triglyceride (mmol/L) MIS Total score (range) Inflammation parameter C-reactive protein (mg/L)
All Patients (n 5 164)
Age ,65 (n 5 106)
Age $65 (n 5 58)
P Value*
58.7 6 12.9 (24.0–85.8) 80/84 58.2 6 50.3 (12.2–269.6)
51.4 6 9.4 (24.0–64.5) 53/53 58.0 6 51.9 (12.2–230.1)
72.0 6 5.6 (65.1–85.8) 27/31 58.6 6 47.7 (12.2–269.6)
,.001 .673 .948
56.5 6 12.3 (31.5–94.7)
58.4 6 12.9 (31.5–94.7)
53.2 6 10.4 (33.4–79.6)
.009
22.3 6 3.6
22.1 6 3.7
22.6 6 3.5
.350
2.9 6 1.1 5.3 6 1.7
3.1 6 1.1 5.4 6 1.6
2.6 6 0.9 4.9 6 1.8
.001 .083
57 67
37 42
20 25
13 27
11 16
2 11
1.3 6 0.2 71.3 6 6.5
1.2 6 0.2 70.5 6 6.4
1.3 6 0.2 72.9 6 6.4
.026 .028
25.7 6 3.2 2.2 6 0.8 18.9 6 2.6
26.1 6 3.5 2.2 6 0.8 19.2 6 2.8
24.9 6 2.7 2.1 6 0.7 18.3 6 2.0
.023 .437 .018
40 6 3 2.0 6 0.4 1.06 6 0.26
41 6 3 2.1 6 0.4 1.11 6 0.26
38 6 3 2.0 6 0.3 0.97 6 0.24
10.8 6 1.4 23.3 6 6.0
11.0 6 1.4 24.5 6 5.9
10.5 6 1.2 21.4 6 5.9
.436
,.001 .036 .001 .023 .002
946 6 204 458 6 89 19 6 13 2.36 6 0.20 1.68 6 0.48 3.97 6 1.23 225 6 262 (8–1,634) 137.3 6 2.9 4.6 6 0.8 4.69 6 1.22 (2.51–11.6)
993 6 204 470 6 89 19 6 13 2.34 6 0.20 1.78 6 0.45 4.13 6 1.19 232 6 252 (8–1,634) 137.1 6 2.8 4.8 6 0.9 4.77 6 1.35 (2.69–11.6)
852 6 160 440 6 83 20 6 13 2.39 6 0.20 1.55 6 0.48 3.68 6 1.26 212 6 280 (8–1,533) 137.8 6 2.9 4.4 6 0.6 4.56 6 0.96 (2.51–7.04)
,.001 .038 .660 .086 .003 .026 .653 .140 ,.001 .304
1.85 6 1.05 (0.43–5.32)
1.90 6 1.12 (0.43–5.32)
1.79 6 0.92 (0.47–4.78)
.546
6.4 6 3.9 (1–22)
5.1 6 3.0 (1–16)
8.7 6 4.3 (2–22)
8.3 6 20.1 (0.1–168.0)
9.0 6 25.1 (0.2–168.0)
7.1 6 6.9 (0.1–34.2)
,.001 .594
IDWG, interdialytic weight gain; RIDWG, relative interdialytic weight gain (IDWG divided by respective dry weight); URR, urea reduction ratio; MAC, mid-arm circumference; TSF, triceps skinfold thickness; MAMC, derived mid-arm muscle circumference (MAMC 5 MAC – [p 3 TSF]); TIBC, total iron-binding capacity; nPCR, normalized protein catabolic rate by volume; BUN, blood urea nitrogen; ALT, alanine aminotransferase; cCa, calcium corrected for serum albumin; P, phosphate; iPTH, intake parathyroid hormone; MIS, Malnutrition Inflammation Score; HD, hemodialysis; ESRD, end-stage renal disease. *For comparison between the two age groups, P values are calculated with Student’s t-test or the chi-square test, as indicated.
214
YANG ET AL
Table 2. Correlation of IDWG/RIDWG With MIS Items in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients
Change in dry weight Dietary intake Gastrointestinal symptoms Functional capacity Comorbidity Decrease or loss of fat Signs of muscle wasting Body mass index Serum albumin Serum TIBC Total score
All Patients (n 5 164)
Age , 65 (n 5 106)
IDWG
RIDWG
IDWG
RIDWG
IDWG
Age $65 (n 5 58) RIDWG
20.203† 20.030 0.040 0.017 20.103 20.072 20.131 20.132 20.127 20.187* 20.157*
20.118 0.173* 0.217† 0.167* 0.045 0.253† 0.160* 0.126 0.049 0.021 0.189*
20.195* 0.005 0.129 0.058 20.082 20.120 20.103 20.220* 20.107 20.260† 20.194*
20.118 0.074 0.138 0.145 0.055 0.162 0.108 0.033 0.090 20.054 0.116
20.122 0.167 0.154 0.219 20.064 0.279* 0.121 0.042 0.083 0.047 0.168
20.069 0.404† 0.430† 0.331* 0.073 0.536‡ 0.416† 0.296* 0.132 0.185 0.467‡
RIDWG, relative interdialytic weight gain; IDWG, interdialytic weight gain; MIS, Malnutrition Inflammation Score; TIBC, total iron-binding capacity. Pearson’s correlation method is used. Shown in each cell is the value of R. The MIS includes scores from 0 to 3, indicating normal to severely abnormal in each component. For detailed definitions of scores 0 to 3 in each component, please refer to the Appendix. *P , .05. †P , .01. ‡P , .001.
middle-aged group; and $65 years (n 5 58), ie, the old group. Compared with young and middle-aged patients, using analysis of variance and a post hoc test with the Bonferroni method, old patients had lower dry weights, IDWGs, MAC, MAMC, serum albumin levels, nPCR, hemoglobin, pre-
dialysis BUN/creatinine, and phosphate and potassium levels, but a higher total MIS (data not shown). The distributions of etiologies of ESRD in the three groups were similar. For conciseness, we later merged the young and middleaged groups. Thus, when performing further analysis, we only divided patients into two
Table 3. Correlation of IDWG/RIDWG With Relevant Parameters in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients (n 5 164)
Delivered dose of dialysis Daugirdas12 Kt/V Anthropometry MAMC Nutritional markers Albumin TIBC nPCR Laboratory data P Na Inflammation parameter C-reactive protein
Age ,65 (n 5 106)
Age $65 (n 5 58)
IDWG
RIDWG
IDWG
RIDWG
IDWG
RIDWG
20.457†
20.111
20.463†
20.099
20.436†
20.072
0.193*
0.074
0.078
0.117
20.001
0.168* 0.286† 0.079
20.009 0.039 0.113
20.054 0.096 0.135
20.093 20.002 20.096
20.118 20.141 20.008
0.170 0.129 0.347† 0.061
0.347† 20.341†
0.316† 20.297†
0.297† 20.344†
0.255† 20.323†
0.331* 20.286*
0.359† 20.227
0.011
0.105
0.010
0.144
20.076
20.060
RIDWG, relative interdialytic weight gain; IDWG, interdialytic weight gain; MAMC, derived mid-arm muscle circumference; TIBC, total iron binding capacity; nPCR, normalized protein catabolic rate by volume; P, phosphate. Pearson’s correlation method is used. Shown in each cell is the value of R. *P , .05. †P , .001.
IDWG AND NUTRITION IN HD
215
Figure 1. Correlations of interdialytic weight gain (IDWG) with serum albumin (upper row), mid-arm muscle circumference (MAMC) (middle row), and body mass index (BMI) (lower row) in all patients (n 5 164; left column), including younger patients (age ,65 years, n 5 106; middle column) and older patients (age $65 years, n 5 58; right column).
age groups: the younger group, aged ,65 years (n 5 106); and the older group, aged $65 years (n 5 58). Comparisons of demographic characteristics and biochemical data between the younger and older groups are given in Table 1.
The age differences in correlations of IDWGs and RIDWGs with the MIS components and other relevant parameters are given in Table 2. When age difference was taken into account, the negative correlations of IDWGs with changes in dry weight and serum TIBC still existed in the
216
YANG ET AL
Figure 2. Correlations of relative interdialytic weight gain (RIDWG) with serum albumin (upper row), mid-arm muscle circumference (MAMC) (middle row), and body mass index (BMI) (lower row) in all patients (n 5 164; left column), including younger patients (age ,65 years, n 5 106; middle column) and older patients (age $65 years, n 5 58; right column).
younger but not the older patient group. On the other hand, the positive correlations of RIDWGs with severity of poor intake, GI symptoms, functional incapacity, and wasting of muscle and subcutaneous fat still existed in the older but not in the younger patient group. Concerning the total
MIS, a negative correlation with IDWG persisted in both patient groups, whereas a positive correlation with RIDWG persisted only in the older group. As shown in Table 3, even where age difference was concerned, IDWGs were also negatively
217
IDWG AND NUTRITION IN HD
correlated with Kt/V in both patient groups; RIDWG was still not significantly correlated with Kt/V. Regarding the relationship with traditional nutritional markers, neither IDWG nor RIDWG had significant correlations with MAMC and serum albumin levels in either patient group, whereas a positive correlation between IDWG and serum TIBC levels persisted in the younger group. In both patient groups, IDWG and RIDWG remained positively correlated with serum phosphate levels, and negatively with serum sodium levels. The correlations of IDWG with serum albumin, MAMC, and BMI in younger and older patients are presented in Figure 1. Those of RIDWG are presented in Figure 2.
Age Differences in Correlations of Dry Weight With MIS Components and Other Relevant Parameters After the foregoing analysis, we found that between IDWG and RIDWG there were discordances in their respective correlations with MIS components and other relevant nutritional markers, especially when age differences were considered. Because RIDWG was calculated as IDWG divided by the respective dry weight, we further analyzed the correlations of dry weight with RIDWG, MIS components, and relevant parameters in the whole cohort, the younger patient group, and the older patient group, respectively. We found that dry weight was negatively correlated with RIDWG in the whole patient group (r 5 –0.168, P 5.031) and the older patient group (r 5 –0.336, P 5.010), but not in the younger patient group. As shown in Table 4, in the whole patient group, dry weight was negatively associated with almost all MIS components except ‘‘change in dry weight.’’ In regard to the differences between the younger and older patient groups, only some of the negative correlations persisted in both groups, whereas some existed only in either group or in neither group. Concerning the correlations of dry weight with other relevant parameters, dry weight was negatively correlated with clearance (Kt/V), regardless of age differences (Table 5). The positive correlations between dry weight and serum albumin and TIBC levels existed only in the younger group, and not in the older group (Table 5). In
Table 4. Correlation of DW With MIS Items in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients Age ,65 Age $65 (n 5 164) (n 5 106) (n 5 58) Change in dry weight Dietary intake Gastrointestinal symptoms Functional capacity Comorbidity Decrease or loss of fat Signs of muscle wasting Body mass index Serum albumin Serum TIBC Total score
20.133
20.120
20.072
20.249† 20.201†
20.051 0.026
20.403† 20.425‡
20.178*
20.090
20.175
20.256‡ 20.438‡
20.240* 20.240 20.389‡ 20.445 ‡
20.415‡
20.318† 20.490‡
20.397‡ 20.229† 20.326‡ 20.484‡
20.404‡ 20.250† 20.337‡ 20.450‡
20.440‡ 20.050 20.260* 20.495‡
MIS, Malnutrition Inflammation Score; TIBC, total ironbinding capacity; DW, dry weight. Pearson’s correlation method is used. Shown in each cell is the value of R. The MIS includes scores from 0 to 3, indicating normal to severely abnormal in each component. For detailed definitions of scores 0 to 3 in each component, please refer to the Appendix. *P , .05. †P , .01. ‡P , .001.
brief, there were also age differences in the correlation of dry weight with MIS components and relevant nutritional markers.
Multivariate Linear Regression of RIDWG and Dry Weight With MIS Components and Other Relevant Parameters To further differentiate the independent effect of RIDWG from that of dry weight, we reanalyzed the corresponding correlations in Tables 2 and 3 using multivariate linear regression, with RIDWG and dry weight regarded as independent variables. The results are presented in Tables 6 and 7. In the whole patient and the younger patient groups, the correlations between RIDWG and the MIS components were almost covered by those of dry weight. On the other hand, in the older patient group, the relationship was still significant between RIDWG and the severity of insufficient food intake, GI upset, functional
218
YANG ET AL
incapacity, and wasting of muscle and subcutaneous fat (Table 6). As shown in Table 7, RIDWG had its own significant correlations, independent of those of dry weight, with delivered dialysis dose and serum levels of phosphate and sodium in the whole, the younger, and the older patient groups.
Table 5. Correlations of DW With Relevant Parameters in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients Age ,65 Age $65 (n 5 164) (n 5 106) (n 5 58) Delivered dose of dialysis Daugirdas12 Kt/V Anthropometry MAMC Nutritional markers Albumin TIBC nPCR Laboratory data P Na Inflammation parameter C-reactive protein
20.639‡
0.201†
20.612‡ 20.670‡
0.159
Discussion
0.218
0.230† 0.382‡ 20.071
0.222* 0.018 0.395‡ 0.256 20.117 20.154
0.116 20.101
0.104 20.088
0.008 20.063
20.101
20.129
20.048
MAMC, derived mid-arm muscle circumference; TIBC, total iron-binding capacity; nPCR, normalized protein catabolic rate by volume; P, phosphate; DW, dry weight. Pearson’s correlation method is used. Shown in each cell is the value of R. *P , .05. †P , .01. ‡P , .001.
In the present study, we found that in younger HD patients, IDWG and RIDWG had almost no significant correlation with MIS components. In older patients, although IDWG also had almost no significant correlation with nutritional parameters, RIDWG had a positive correlation with severity of insufficient food intake, GI upset, functional incapacity, and wasting of muscle and subcutaneous fat (Table 2). Our findings suggest that in older HD patients, the greater the RIDWG, the poorer the nutritional status will be. On the other hand, we also found that dry weight was negatively correlated with severity of insufficient food intake and GI upset in older HD patients (Table 4). Because dry weight was the denominator and negatively correlated with RIDWG in older HD patients, the associations of higher RIDWG with higher MIS component
Table 6. Multivariate Linear Regression of RIDWG and DW With MIS Items in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients (n 5 164)
Change in dry weight Dietary intake Gastrointestinal symptoms Functional capacity Comorbidity Decrease or loss of fat Signs of muscle wasting Body mass index Serum albumin Serum TIBC Total score
Age ,65 (n 5 106)
Age $65 (n 5 58)
RIDWG
DW
RIDWG
DW
RIDWG
DW
20.145 0.135 0.188* 0.141 0.003 0.184† 0.092 0.061 0.011 20.035 0.111
20.158* 20.226† 20.169* 20.154* 20.255† 20.407‡ 20.399‡ 20.386‡ 20.227† 20.332‡ 20.466‡
20.137 0.068 0.145 0.135 0.022 0.110 0.065 20.024 0.056 20.103 0.055
20.139 20.041 0.046 20.072 20.237* 20.374‡ 20.309† 20.407‡ 20.242* 20.352‡ 20.443‡
20.106 0.303* 0.324* 0.307* 20.009 0.436‡ 0.283* 0.167 0.130 0.110 0.339†
20.108 20.302* 20.316* 20.072 20.243 20.298* 20.395† 20.384† 20.006 20.223 20.382†
RIDWG, relative interdialytic weight gain; MIS, Malnutrition Inflammation Score; TIBC, total iron-binding capacity; DW, dry weight. The RIDWG and DW are entered as independent variables, and other relevant parameters are entered as dependent variables. Shown in each cell is the value of the standardized coefficient (b). The MIS includes scores from 0 to 3, indicating normal to severely abnormal in each component. For detailed definitions of scores 0 to 3 in each component, please refer to the Appendix. *P , .05. †P , .01. ‡P , .001.
219
IDWG AND NUTRITION IN HD
Table 7. Multivariate Linear Regression of RIDWG and DW With Relevant Parameters in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients (n 5 164)
Delivered dose of dialysis Daugirdas12 Kt/V Anthropometry MAMC Nutritional markers Albumin TIBC nPCR Laboratory data P Na Inflammation parameter C-reactive protein
Age ,65 (n 5 106)
Age $65 (n 5 58)
RIDWG
DW
RIDWG
DW
RIDWG
DW
20.237‡
20.683‡
20.206†
20.646‡
20.338†
20.784‡
0.173
0.082
0.246
0.112
0.220†
0.102
0.031 0.105 0.104
0.235† 0.399‡ 20.052
20.023 0.154 0.119
0.219* 0.417‡ 20.099
20.126 20.063 20.068
20.025 0.235 20.177
0.345‡ 20.323‡
0.174* 20.155*
0.275† 20.342‡
0.142 0.135
0.408† 20.280*
0.145 20.157
0.091
20.085
0.128
20.084
20.075
20.110
RIDWG, relative interdialytic weight gain; MAMC, derived mid-arm muscle circumference; TIBC, total iron-binding capacity; nPCR, normalized protein catabolic rate by volume; P, phosphate; DW, dry weight. The RIDWG and dry weight are entered as independent variables and other relevant parameters as dependent variables. Shown in each cell is the value of standardizied coefficient (b). *P , .05. †P , .01. ‡P , .001.
scores of food intake and GI upset would only be the reciprocal reflection of the associations between dry weight and MIS component scores. It is reasonable that older HD patients with insufficient food intake and more GI upset would have a lower dry weight and, consequently, higher RIDWG. However, as shown in Table 6, in the older patient group, the correlations of RIDWG were independent from those of dry weight, concerning severity of insufficient food intake, GI upset, functional incapacity, and wasting of muscle and subcutaneous fat. In regard to the discordances between IDWG and RIDWG in their correlations with MIS components, especially when age differences are considered, there are several possible explanations. First, younger patients would have ingested a more than adequate amount of food to maintain an acceptable nutritional status. Therefore, IDWG lost its correlation with other nutritional markers in younger HD patients. Second, as previously mentioned, the age differences of dry weight in the corresponding correlations with MIS components would be responsible to a certain extent. Third, the age difference in dietary intake could also be contributory. Regarding dietary intake status in the younger patient group (n 5 106), 93 patients had a score of 0, and 13 had a score of 1; in
the older patient group (n 5 58), 33 patients had a score of 0, 22 had a score of 1, and 3 had a score of 2. Thus the older patients had a higher rate of ingesting semifluid and full-fluid foods than did younger patients (c2 5 21.695, P , .001), compatible with the higher rate of dental problems in older patients. Fourth, malnourished older patients would have taken more liquid nutritional supplements,13 such as Nepro (Abbott Laboratories Services Corp., Columbus, OH). Fifth, malnourished older patients would have a higher rate of ingesting semifluid or full-fluid food via the nasogastric tube. Some studies found that IDWG was correlated negatively with patient age6,14,15 and positively with serum albumin levels6–8 in HD patients. Consequently, ignoring the confounding effect that younger patients actually have better nutritional status, IDWG suggested as a positive nutritional marker.16 Nevertheless, taking age differences into consideration is a logical first step, while investigating whether IDWG can be regarded as a nutritional marker warrants further study. In the present study, IDWG was correlated positively with serum albumin levels in the whole patient group (Table 3). However, the positive correlation did not exist in the younger or the older patient groups (Table 3). Therefore, in accordance with our previous
220
YANG ET AL
proposal, it is hard to present IDWG as a positive nutritional marker when patient age is considered. With regard to the relationship between IDWG and nPCR, a positive correlation was reported.8,14,15 However, there was no similar finding in the present study. Taking age difference into consideration, we found that serum albumin levels positively correlated with nPCR in the whole (r 5 0.289, P ,.001) and in the older (r 5 0.428, P 5 .001) patient groups, but not in the younger group (r 5 0.089, P 5.366). That is, the positive correlation between nPCR and albumin levels only existed in the older patient group. Therefore, we suggest that the results of previous studies concerning the relationships among nutritional markers be reexamined, taking age difference into consideration. Regardless of age difference, IDWG has a negative correlation with dialysis clearance (Daugirdas Kt/V), although RIDWG does not (Table 3). A similar finding was reported by Lopez-Gomez et al.6 If the confounding effects of dry weight shown in Table 5 are taken into account, the negative correlations of IDWGs would be offset by the negative correlations of dry weight. Therefore, it is not surprising that RIDWG has no significant correlation with dialysis clearance. Before the results of further study are made available, we assume that the negative correlation between IDWG and dialysis clearance can be partly, if not completely, explained by a lower average blood flow attributable to more episodes of interdialytic hypotension. In clinical practice, higher IDWG and RIDWG cause intradialytic hypotension more frequently.17,18 Logically, IDWG is assumed to have a positive correlation with the amount of food intake in HD patients,19,20 especially when patients become anuric after at least 1 year of HD therapy. The RIDWG is also assumed to correlate positively with the relative amounts of food intake in HD patients when individual dry weights are considered. The above inferences are indirectly supported by the findings of our study. As shown in Table 3, IDWG and RIDWG actually have a significantly positive correlation with serum phosphate levels (a positive indicator of food intake),21 and a negative correlation with serum sodium levels (a negative surrogate of sodium-free water intake). Lopez-Gomez JM et al reported similar findings.6 Furthermore, as seen in Table 1, the older patient group had significantly less IDWG and dry weight than the younger group. Meanwhile, there was
only a marginal difference in RIDWG (Table 1). Therefore, in accordance with previous studies,19,20 we agree that younger HD patients demonstrate more food intake than older patients. However, the relative amounts of food intake regarding respective dry weight would be similar in both groups of patients. In conclusion, we demonstrated the indispensable role of age difference when analyzing relationships among nutritional markers. Nevertheless, our findings here do not fully support the contention that IDWG reflects caloric and protein intake and can be regarded as a nutritional marker, as suggested in previous studies. Although IDWG and RIDWG could be positively correlated with the amount of food intake, we also found that RIDWG would be a negative nutritional maker, correlating with MIS in older HD patients. Perhaps high IDWG is associated with high salt and water intake rather than caloric and protein intake. This might especially be an issue in this study as this appears to be conducted in an Asian population. Chinese food is high in salt. Therefore, until more clinical evidence becomes available to differentiate food intake from nutritional gain, it remains controversial to regard IDWG and RIDWG as nutritional markers, especially when age difference is considered.
References 1. Sharp J, Wild MR, Gumley AI, Deighan CJ: A cognitive behavioral group approach to enhance adherence to hemodialysis fluid restrictions: A randomized controlled trial. Am J Kidney Dis 45:1046-1057, 2005 2. Sharp J, Wild MR, Gumley AI: A systematic review of psychological interventions for the treatment of nonadherence to fluid-intake restrictions in people receiving hemodialysis. Am J Kidney Dis 45:15-27, 2005 3. Foley RN, Herzog CA, Collins AJ: United States Renal Data System. Blood pressure and long-term mortality in United States hemodialysis patients: USRDS Waves 3 and 4 Study. Kidney Int 62:1784-1790, 2002 4. Hecking E, Bragg-Gresham JL, Rayner HC, et al: Haemodialysis prescription, adherence and nutritional indicators in five European countries: Results from the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrol Dial Transplant 19: 100-107, 2004 5. Rahman M, Fu P, Sehgal AR, Smith MC: Interdialytic weight gain, compliance with dialysis regimen, and age are independent predictors of blood pressure in hemodialysis patients. Am J Kidney Dis 35:257-265, 2000 6. Lopez-Gomez JM, Villaverde M, Jofre R, RodriguezBenitez P, Perez-Garcia R: Interdialytic weight gain as a marker of blood pressure, nutrition, and survival in hemodialysis patients. Kidney Int 93(suppl):S63-S68, 2005
IDWG AND NUTRITION IN HD 7. Szczech LA, Reddan DN, Klassen PS, et al: Interactions between dialysis-related volume exposures, nutritional surrogates and mortality among ESRD patients. Nephrol Dial Transplant 18: 1585-1591, 2003 8. Testa A, Beaud JM: The other side of the coin: interdialytic weight gain as an index of good nutrition. Am J Kidney Dis 31: 830-834, 1998 9. Enia G, Sicuso C, Alati G, et al: Subjective global assessment of nutrition in dialysis patients. Nephrol Dial Transplant 8: 1094-2008, 1993 10. Kalantar-Zadeh K, Kopple JD, Block G, et al: A malnutrition-inflammation score is correlated with morbidity and mortality in maintenance hemodialysis patients. Am J Kidney Dis 38: 1251-1263, 2001 11. Morduchowicz G, Winkler J, Zabludowski JR, Boner G: Effects of residual renal function in haemodialysis patients. Int Urol Nephrol 26:125-131, 1994 12. Daugirdas JT: Second generation logarithmic estimates of single-pool variable volume Kt/V: An analysis of error. J Am Soc Nephrol 4:1205-1213, 1993 13. Caglar K, Fedje L, Dimmitt R, et al: Therapeutic effects of oral nutritional supplementation during hemodialysis. Kidney Int 62:1054-1059, 2002 14. Kimmel PL, Varela MP, Peterson RA, et al: Interdialytic weight gain and survival in hemodialysis patients: Effects of duration of ESRD and diabetes mellitus. Kidney Int 57:1141-1151, 2000 15. Ifudu O, Uribarri J, Rajwani I, et al: Relation between interdialytic weight gain, body weight and nutrition in hemodialysis patients. Am J Nephrol 22:363-368, 2002 16. Sarkar SR, Kotanko P, Levin NW: Interdialytic weight gain: Implications in hemodialysis patients. Semin Dial 19:429-433, 2006 17. Takeda A, Toda T, Fujii T, Sasaki S, Matsui N: Can predialysis hypertension prevent intradialytic hypotension in hemodialysis patients? Nephron 103:c137-c143, 2006 18. Kursat S, Ozgur B, Alici T: Effect of ultrafiltration on blood pressure variability in hemodialysis patients. Clin Nephrol 59: 289-292, 2003 19. Testa A, Plou A: Clinical determinants of interdialytic weight gain. J Ren Nutr 11:155-160, 2001 20. Bossola M, Muscaritoli M, Tazza L, et al: Variables associated with reduced dietary intake in hemodialysis patients. J Ren Nutr 15:244-252, 2005 21. Levy J, Morgan J, Brown E: Assessment of nutitional status: biochemical. In: Levy J, Morgan J, Brown E (eds.): Oxford Handbook of Dialysis. New York: Oxford University Press, Inc., 2001, pp 388-389
Appendix
The following are details of the definitions of each component of the Malnutrition Inflammation Score copied from the original proposal of Kalantar-Zadeh et al.10 Weight change is determined as the change in edema-free postdialysis body weight in the previous 6 months. The lowest score (0) is given if weight loss is ,0.5 kg, or there is an increase in body weight. A score of 1 indicates a minor loss of at least 0.5 kg, but ,1.0 kg. A score of 2 is given for weight loss of at least 1.0 kg, but ,5% of body weight, and a score of 3 indicates a weight
221
loss of $5%. Dietary intake is scored as 0 if it involves the usual intake of solid foods, with no recent decrease in the amount or quality of meals. A score of 1 indicates a slightly suboptimal solid diet, a score of 2 indicates a full-liquid diet or moderate decrease in food intake, and a score of 3 indicates a daily nutrient intake that would be incompatible with life on a long-term basis. Gastrointestinal symptoms are scored as 0 if the patient has a good appetite and no GI symptoms. A score of 1 indicates mildly decreased appetite or mild nausea. A score of 2 indicates occasional vomiting or other moderate GI symptoms, such as abdominal pain. A score of 3 indicates diarrhea, frequent vomiting, or severe anorexia. Functional capacity is scored as 0 for normal functional capacity or a considerable improvement in the level of previous functional impairment. A score of 1 indicates mild or occasional difficulty with baseline ambulation or feeling tired frequently. A score of 2 indicates difficulty with independent activities. A score of 3 indicates restriction to light activity or a persistent bedbound or chairbound state. Comorbidity is scored as 0 if there are no other medical illnesses and the patient has undergone hemodialysis therapy for ,1 year; as 1 for mild comorbidity, excluding such major comorbid conditions (MCCs) as congestive heart failure class III or IV, severe coronary artery disease, clinically evident acquired immunodeficiency syndrome, moderate to severe chronic obstructive pulmonary disease, and metastatic malignancies, or dialysis therapy for 1 to 4 years; as 2 for moderate comorbidity (including one of the diseases listed under MCCs) or dialysis therapy for .4 years; and as 3 for $2 MCCs. The existence of diabetes per se is not accounted for if the previously mentioned comorbidities do not exist. Instead, comorbidities that may pose a risk for poor outcomes in patients with diabetes are examined individually. Body-fat stores are scored by assessing subcutaneous fat deposition in four body areas: below the eyes, the triceps, the biceps, and the chest. Signs of muscle wasting are determined by briefly examining seven sites: the temple, clavicle, scapula, ribs (intercostal spaces), quadriceps, knee, and interosseous muscles. For each of these two components, a score of 0 to 3, representing normal to severe changes, is assigned according to conventional SGA guidelines. Body mass index was graded at four levels (0 to 3), representing BMI .20, 18 to 19.99, 16 to 17.99, and ,16 kg/m2, respectively. A score of 0
222
YANG ET AL
in serum albumin means that the level is $40 g/L (or $4.0 g/dL). A score of 1 indicates 35 to 39 g/L (or 3.5 to 3.9 g/dL). A score of 2 indicates 30 to 34 g/L (or 3.0 to 3.4 g/dL). A score of score 3 indicates ,30 g/L (or ,3.0 g/dL).
The scores 0 to 3 in TIBC represent TIBC $2.50 g/L (or 250 mg/dL), 2.00 to 2.49 g/L (or 200 to 249 mg/dL), 1.50 to1.99 g/L (or 150 to 199 mg/dL), and ,1.50 g/L (or ,150 mg/dL), respectively.
P
ATIENTS UNDERGOING maintenance hemodialysis (HD) are always advised to control their fluid intake strictly in order to prevent long-term adverse cardiovascular outcomes and intradialytic hypotension because of large ultrafiltration.1–5 However, out of fear that inadequate food intake will worsen nutritional status, some patients always have greater interdialytic weight *Department of Nursing, Far Eastern Memorial Hospital, Taipei, Taiwan. †Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan. ‡Department of Internal Medicine, Far Eastern Memorial Hospital, Taipei, Taiwan. S.-C.Y. and C.-K.C. contributed equally to this article. Address reprint requests to Shih-Ping Hsu, MD, Department of Internal Medicine, Far Eastern Memorial Hospital, No. 21, Sec. 2, Nan-Ya South Road, Pan-Chiao, Taipei 244, Taiwan.
E-mail: [email protected] 2008 by the National Kidney Foundation, Inc. 1051-2276/08/1802-0006$34.00/0 doi:10.1053/j.jrn.2007.11.012
Ó
210
gain (IDWG) than advised. Conversely, other patients with less of an IDWG actually have inadequate food intake and poor nutritional status. Therefore, it is clinically interesting and important to determine whether IDWG reflects caloric and protein intake and correlates with other nutritional markers in HD patients. In the literature, some studies showed that IDWG is positively correlated with serum albumin levels in HD patients.6–8 Furthermore, it was proposed that HD patients with a large IDWG have better survival outcomes.6 Yet the relationship between IDWG and subjective global assessment (SGA)9 parameters has not been well-discussed. Thus, in this cross-sectional study, we attempted to investigate correlations of IDWG with SGA parameters and other nutritional markers in chronic HD patients, taking age differences into particular account. For comprehensive coverage, the Malnutrition Inflammation Score (MIS),10 an extended and modified form of SGA, was adopted. Journal of Renal Nutrition, Vol 18, No 2 (March), 2008: pp 210–222
IDWG AND NUTRITION IN HD
Subjects and Methods Patients To minimize the confounding effects of residual renal function on IDWG and nutritional markers,11 we enrolled only anuric (,50 mL/ day) HD patients who had undergone HD three times a week for .12 months at the outpatient HD unit of the Far Eastern Memorial Hospital (Taipei, Taiwan). Considering known confounding factors for nutritional status, we excluded patients with the following conditions or events within the preceding 3 months: obvious infections or inflammatory diseases, hospitalizations, and major surgery. In total, 164 patients were enrolled. This clinical study adhered to the Declaration of Helsinki, and informed consent was obtained from patients. Regarding the demographics and etiology of end-stage renal disease (ESRD), we obtained enrollment data from the medical charts and computerized database at the HD unit. Laboratory Parameters All biochemical analyses were performed with a Hitachi 747 Automatic Analyzer (Hitachi, Tokyo, Japan). The results of routine monthly midweek, nonfasting, predialysis blood samples were used for analysis. Postdialysis blood urea nitrogen (BUN) levels of the same dialysis session were measured to calculate the delivered dose of dialysis (Kt/V) by the method of Daugirdas.12 The predialysis BUN level of the next dialysis was measured to calculate the normalized protein catabolic rate (nPCR, normalized to body weight derived from the urea distribution space [Vurea/ 0.58]). Serum calcium levels were corrected for serum albumin, according to the following formula: collected serum Ca (mmol/L) 5 serum calcium (mmol/L) 1 0.02 3 [40 – serum albumin (g/L)]. Dialysis Therapy All patients had undergone HD three times a week with bicarbonate-based dialysate, volumetric ultrafiltration control, and single-use, cellulose-based membranes. The dry weight of each patient was determined on a trial-and-error basis and reevaluated frequently. Each patient was visited by our nephrologists at each HD session. The session time lasted 4.0 6 0.3 hours (range, 3.0 to 4.5 hours). The prescriptions of recombi-
211
nant Human Erythropoietin and iron supplement were based on a standardized algorithm, with the target Hematocrit at 30% to 36%.
Interdialytic Weight Gain and Relative Interdialytic Weight Gain With regard to IDWG, the average of all 12 sessions within the previous 4 weeks before enrollment was used for analysis. The relative IDWG (RIDWG) was calculated as the average IDWG divided by the respective dry weight. Malnutrition Inflammation Score (Subjective Global Assessment) Proposed by Kalantar-Zadeh et al10 as a comprehensive, risk-stratifying scoring system for morbidity and mortality in maintenance HD patients, the Malnutrition Inflammation Score (MIS) integrates patients’ nutritional history, physical examination findings, body mass index, and laboratory parameters. Concerning patients’ nutritional history, the following five components, adopted from the original subjective global assessment (SGA),9 were taken into account: overall change in end-dialysis dry weight in the past 5 months, dietary intake, gastrointestinal (GI) symptoms, impairment of nutritionally related functional capacity, and comorbidity, including number of years on dialysis. Based on the SGA criteria, physical examinations involved two observations, ie, decreased fat stores or loss of subcutaneous fat, and signs of muscle-wasting. Body mass index (BMI) was calculated as end-dialysis body weight in kilograms divided by the square of body height in meters. Serum albumin levels and total iron-binding capacity (TIBC) were selected as the pertinent and representative laboratory parameters. Thus, the MIS has a total of 10 components, each with four levels of severity, from 0 (normal) to 3 (severely abnormal). The sum of all 10 MIS components ranges from 0 (normal) to 30 (severely malnourished). A higher score reflects a more severe degree of malnutrition and inflammation. Through the courtesy of K. Kalantar-Zadeh, detailed definitions of each component of MIS have been supplied as an Appendix. With reference to the original scoring method proposed by Kalantar-Zadeh et al,10 we assigned a trained physician’s assistant to score each patient within 5 to 15 minutes before anthropometric
212
YANG ET AL
measurements were performed. To evaluate the degree of reproducibility, the same physician’s assistant repeated the MIS assessment after 1 week on a subset of 20 patients without reference to their first MIS evaluation. Compared with the findings in the previous report, the correlation coefficient (r) between the two MIS assessments was 0.90, denoting a good degree of reproducibility.
Anthropometric Evaluation Another trained physician’s assistant performed anthropometric measurements 5 to 20 minutes after the termination of an HD session. These measurements included triceps skinfold thickness (TSF), mid-arm circumference (MAC), and derived mid-arm muscle circumference (MAMC). All anthropometric measurements were performed by the trained physician’s assistant three times in rapid succession on the nonaccess-containing arm of each dialysis patient, and the average of the three measurements was regarded as the final result for analysis.
main causes of ESRD. In these patients, the IDWG was 2.9 6 1.1 kg, and the RIDWG was 5.3% 6 1.7%. In almost all components of the MIS, most of the patients scored 0, ie, normal. The exception was the component of ‘‘comorbidity’’ in which the score was at least 1, because we only enrolled patients who had undergone HD for .1 year. The sum of the MIS ranged from 1 to 22, with a mean of 6.4 6 3.9.
Results
Relationship of IDWG and RIDWG With Severity of Individual Components of MIS and Other Relevant Parameters At first, the relationship of IDWG and RIDWG with the severity of each MIS component and other relevant parameters was analyzed in the group of all enrolled patients (n 5 164). As shown in Table 2, IDWG was negatively correlated with change in dry weight and serum TIBC levels, whereas RIDWG was positively correlated with the severity of poor intake, GI symptoms, functional incapacity, and wasting of muscle and subcutaneous fat. With the sum of the MIS, IDWG had a negative correlation, whereas RIDWG paradoxically had a positive correlation. As shown in Table 3, whereas IDWG was negatively correlated with dialysis clearance (Kt/V), RIDWG was not significantly correlated with Kt/V. Regarding the traditional nutritional markers, MAMC, serum levels of albumin, and TIBC were positively correlated with IDWG, but were not significantly correlated with RIDWG. On the other hand, both IDWG and RIDWG were positively correlated with serum phosphate levels, and negatively correlated with serum sodium levels. The correlations of IDWG with serum albumin, MAMC, and BMI in all patients are shown in Figure 1. Those of RIDWG are shown in Figure 2.
Characteristics of Patients and Laboratory Data The demographic characteristics, dialysis parameters, and biochemical data of patients are presented in Table 1. Of 164 patients, 84 (51.2%) were women, and 80 (48.8%) were men. Their age was 58.7 6 12.9 years (range, 24.0 to 85.8 years). The duration of HD therapy ranged from 12.2 to 269.6 months. Diabetic nephropathy (34.8%) and chronic glomerulonephritis (40.9%) were the
Age Differences in the Correlation of IDWG and RIDWG With MIS Components and Other Relevant Parameters To determine age differences in the correlation of IDWG and RIDWG with MIS components and other relevant parameters, we arbitrarily divided the enrolled patients (n 5 164) into three age groups: #45 years (n 5 22), ie, the young group; between 45 and 65 years (n 5 84), ie, the
Statistical Analysis Unless otherwise stated, continuous variables are presented as mean values 6 standard deviation (SD), and categorical variables as a number for each item. For comparing differences between age groups, Student’s t-test and the chi-square method were used as indicated. For evaluating correlations between IDWG and RIDWG and the severity of individual components in the MIS, the Pearson correlation method was used. A two-tailed P , .05 was considered statistically significant, and a P value of between .05 and .10 was considered marginally significant. All computations were performed with SPSS 10.0 for Windows (SPSS, Inc., Chicago, IL).
213
IDWG AND NUTRITION IN HD
Table 1. Demographic Characteristics and Biochemical Data of All Patients, Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients Variables Demographics Age (y) (range, y) Sex (male/female) Duration on HD (mo) (range, mo) Dry weight (kg) (range, kg) Body mass index (kg/m2) IDWG (kg) RIDWG (% of dry weight) Etiology of ESRD Diabetic nephropathy Chronic glomerulonephritis Others Undetermined Delivered dose of dialysis Daugirdas12 Kt/V URR (%) Anthropometry MAC (cm) TSF (cm) MAMC (cm) Nutrition Albumin (g/L) TIBC (g/L) nPCR (g/kg/day) Laboratory data Hemoglobin (g/dL) Predialysis BUN (mmol/L) Creatinine (mmol/L) Uric acid (mmol/L) ALT (IU/L) cCa (mmol/L) P (mmol/L) cCa 3 P iPTH (ng/L) (range) Na (meq/L) K (meq/L) Total cholesterol (mmol/L) Triglyceride (mmol/L) MIS Total score (range) Inflammation parameter C-reactive protein (mg/L)
All Patients (n 5 164)
Age ,65 (n 5 106)
Age $65 (n 5 58)
P Value*
58.7 6 12.9 (24.0–85.8) 80/84 58.2 6 50.3 (12.2–269.6)
51.4 6 9.4 (24.0–64.5) 53/53 58.0 6 51.9 (12.2–230.1)
72.0 6 5.6 (65.1–85.8) 27/31 58.6 6 47.7 (12.2–269.6)
,.001 .673 .948
56.5 6 12.3 (31.5–94.7)
58.4 6 12.9 (31.5–94.7)
53.2 6 10.4 (33.4–79.6)
.009
22.3 6 3.6
22.1 6 3.7
22.6 6 3.5
.350
2.9 6 1.1 5.3 6 1.7
3.1 6 1.1 5.4 6 1.6
2.6 6 0.9 4.9 6 1.8
.001 .083
57 67
37 42
20 25
13 27
11 16
2 11
1.3 6 0.2 71.3 6 6.5
1.2 6 0.2 70.5 6 6.4
1.3 6 0.2 72.9 6 6.4
.026 .028
25.7 6 3.2 2.2 6 0.8 18.9 6 2.6
26.1 6 3.5 2.2 6 0.8 19.2 6 2.8
24.9 6 2.7 2.1 6 0.7 18.3 6 2.0
.023 .437 .018
40 6 3 2.0 6 0.4 1.06 6 0.26
41 6 3 2.1 6 0.4 1.11 6 0.26
38 6 3 2.0 6 0.3 0.97 6 0.24
10.8 6 1.4 23.3 6 6.0
11.0 6 1.4 24.5 6 5.9
10.5 6 1.2 21.4 6 5.9
.436
,.001 .036 .001 .023 .002
946 6 204 458 6 89 19 6 13 2.36 6 0.20 1.68 6 0.48 3.97 6 1.23 225 6 262 (8–1,634) 137.3 6 2.9 4.6 6 0.8 4.69 6 1.22 (2.51–11.6)
993 6 204 470 6 89 19 6 13 2.34 6 0.20 1.78 6 0.45 4.13 6 1.19 232 6 252 (8–1,634) 137.1 6 2.8 4.8 6 0.9 4.77 6 1.35 (2.69–11.6)
852 6 160 440 6 83 20 6 13 2.39 6 0.20 1.55 6 0.48 3.68 6 1.26 212 6 280 (8–1,533) 137.8 6 2.9 4.4 6 0.6 4.56 6 0.96 (2.51–7.04)
,.001 .038 .660 .086 .003 .026 .653 .140 ,.001 .304
1.85 6 1.05 (0.43–5.32)
1.90 6 1.12 (0.43–5.32)
1.79 6 0.92 (0.47–4.78)
.546
6.4 6 3.9 (1–22)
5.1 6 3.0 (1–16)
8.7 6 4.3 (2–22)
8.3 6 20.1 (0.1–168.0)
9.0 6 25.1 (0.2–168.0)
7.1 6 6.9 (0.1–34.2)
,.001 .594
IDWG, interdialytic weight gain; RIDWG, relative interdialytic weight gain (IDWG divided by respective dry weight); URR, urea reduction ratio; MAC, mid-arm circumference; TSF, triceps skinfold thickness; MAMC, derived mid-arm muscle circumference (MAMC 5 MAC – [p 3 TSF]); TIBC, total iron-binding capacity; nPCR, normalized protein catabolic rate by volume; BUN, blood urea nitrogen; ALT, alanine aminotransferase; cCa, calcium corrected for serum albumin; P, phosphate; iPTH, intake parathyroid hormone; MIS, Malnutrition Inflammation Score; HD, hemodialysis; ESRD, end-stage renal disease. *For comparison between the two age groups, P values are calculated with Student’s t-test or the chi-square test, as indicated.
214
YANG ET AL
Table 2. Correlation of IDWG/RIDWG With MIS Items in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients
Change in dry weight Dietary intake Gastrointestinal symptoms Functional capacity Comorbidity Decrease or loss of fat Signs of muscle wasting Body mass index Serum albumin Serum TIBC Total score
All Patients (n 5 164)
Age , 65 (n 5 106)
IDWG
RIDWG
IDWG
RIDWG
IDWG
Age $65 (n 5 58) RIDWG
20.203† 20.030 0.040 0.017 20.103 20.072 20.131 20.132 20.127 20.187* 20.157*
20.118 0.173* 0.217† 0.167* 0.045 0.253† 0.160* 0.126 0.049 0.021 0.189*
20.195* 0.005 0.129 0.058 20.082 20.120 20.103 20.220* 20.107 20.260† 20.194*
20.118 0.074 0.138 0.145 0.055 0.162 0.108 0.033 0.090 20.054 0.116
20.122 0.167 0.154 0.219 20.064 0.279* 0.121 0.042 0.083 0.047 0.168
20.069 0.404† 0.430† 0.331* 0.073 0.536‡ 0.416† 0.296* 0.132 0.185 0.467‡
RIDWG, relative interdialytic weight gain; IDWG, interdialytic weight gain; MIS, Malnutrition Inflammation Score; TIBC, total iron-binding capacity. Pearson’s correlation method is used. Shown in each cell is the value of R. The MIS includes scores from 0 to 3, indicating normal to severely abnormal in each component. For detailed definitions of scores 0 to 3 in each component, please refer to the Appendix. *P , .05. †P , .01. ‡P , .001.
middle-aged group; and $65 years (n 5 58), ie, the old group. Compared with young and middle-aged patients, using analysis of variance and a post hoc test with the Bonferroni method, old patients had lower dry weights, IDWGs, MAC, MAMC, serum albumin levels, nPCR, hemoglobin, pre-
dialysis BUN/creatinine, and phosphate and potassium levels, but a higher total MIS (data not shown). The distributions of etiologies of ESRD in the three groups were similar. For conciseness, we later merged the young and middleaged groups. Thus, when performing further analysis, we only divided patients into two
Table 3. Correlation of IDWG/RIDWG With Relevant Parameters in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients (n 5 164)
Delivered dose of dialysis Daugirdas12 Kt/V Anthropometry MAMC Nutritional markers Albumin TIBC nPCR Laboratory data P Na Inflammation parameter C-reactive protein
Age ,65 (n 5 106)
Age $65 (n 5 58)
IDWG
RIDWG
IDWG
RIDWG
IDWG
RIDWG
20.457†
20.111
20.463†
20.099
20.436†
20.072
0.193*
0.074
0.078
0.117
20.001
0.168* 0.286† 0.079
20.009 0.039 0.113
20.054 0.096 0.135
20.093 20.002 20.096
20.118 20.141 20.008
0.170 0.129 0.347† 0.061
0.347† 20.341†
0.316† 20.297†
0.297† 20.344†
0.255† 20.323†
0.331* 20.286*
0.359† 20.227
0.011
0.105
0.010
0.144
20.076
20.060
RIDWG, relative interdialytic weight gain; IDWG, interdialytic weight gain; MAMC, derived mid-arm muscle circumference; TIBC, total iron binding capacity; nPCR, normalized protein catabolic rate by volume; P, phosphate. Pearson’s correlation method is used. Shown in each cell is the value of R. *P , .05. †P , .001.
IDWG AND NUTRITION IN HD
215
Figure 1. Correlations of interdialytic weight gain (IDWG) with serum albumin (upper row), mid-arm muscle circumference (MAMC) (middle row), and body mass index (BMI) (lower row) in all patients (n 5 164; left column), including younger patients (age ,65 years, n 5 106; middle column) and older patients (age $65 years, n 5 58; right column).
age groups: the younger group, aged ,65 years (n 5 106); and the older group, aged $65 years (n 5 58). Comparisons of demographic characteristics and biochemical data between the younger and older groups are given in Table 1.
The age differences in correlations of IDWGs and RIDWGs with the MIS components and other relevant parameters are given in Table 2. When age difference was taken into account, the negative correlations of IDWGs with changes in dry weight and serum TIBC still existed in the
216
YANG ET AL
Figure 2. Correlations of relative interdialytic weight gain (RIDWG) with serum albumin (upper row), mid-arm muscle circumference (MAMC) (middle row), and body mass index (BMI) (lower row) in all patients (n 5 164; left column), including younger patients (age ,65 years, n 5 106; middle column) and older patients (age $65 years, n 5 58; right column).
younger but not the older patient group. On the other hand, the positive correlations of RIDWGs with severity of poor intake, GI symptoms, functional incapacity, and wasting of muscle and subcutaneous fat still existed in the older but not in the younger patient group. Concerning the total
MIS, a negative correlation with IDWG persisted in both patient groups, whereas a positive correlation with RIDWG persisted only in the older group. As shown in Table 3, even where age difference was concerned, IDWGs were also negatively
217
IDWG AND NUTRITION IN HD
correlated with Kt/V in both patient groups; RIDWG was still not significantly correlated with Kt/V. Regarding the relationship with traditional nutritional markers, neither IDWG nor RIDWG had significant correlations with MAMC and serum albumin levels in either patient group, whereas a positive correlation between IDWG and serum TIBC levels persisted in the younger group. In both patient groups, IDWG and RIDWG remained positively correlated with serum phosphate levels, and negatively with serum sodium levels. The correlations of IDWG with serum albumin, MAMC, and BMI in younger and older patients are presented in Figure 1. Those of RIDWG are presented in Figure 2.
Age Differences in Correlations of Dry Weight With MIS Components and Other Relevant Parameters After the foregoing analysis, we found that between IDWG and RIDWG there were discordances in their respective correlations with MIS components and other relevant nutritional markers, especially when age differences were considered. Because RIDWG was calculated as IDWG divided by the respective dry weight, we further analyzed the correlations of dry weight with RIDWG, MIS components, and relevant parameters in the whole cohort, the younger patient group, and the older patient group, respectively. We found that dry weight was negatively correlated with RIDWG in the whole patient group (r 5 –0.168, P 5.031) and the older patient group (r 5 –0.336, P 5.010), but not in the younger patient group. As shown in Table 4, in the whole patient group, dry weight was negatively associated with almost all MIS components except ‘‘change in dry weight.’’ In regard to the differences between the younger and older patient groups, only some of the negative correlations persisted in both groups, whereas some existed only in either group or in neither group. Concerning the correlations of dry weight with other relevant parameters, dry weight was negatively correlated with clearance (Kt/V), regardless of age differences (Table 5). The positive correlations between dry weight and serum albumin and TIBC levels existed only in the younger group, and not in the older group (Table 5). In
Table 4. Correlation of DW With MIS Items in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients Age ,65 Age $65 (n 5 164) (n 5 106) (n 5 58) Change in dry weight Dietary intake Gastrointestinal symptoms Functional capacity Comorbidity Decrease or loss of fat Signs of muscle wasting Body mass index Serum albumin Serum TIBC Total score
20.133
20.120
20.072
20.249† 20.201†
20.051 0.026
20.403† 20.425‡
20.178*
20.090
20.175
20.256‡ 20.438‡
20.240* 20.240 20.389‡ 20.445 ‡
20.415‡
20.318† 20.490‡
20.397‡ 20.229† 20.326‡ 20.484‡
20.404‡ 20.250† 20.337‡ 20.450‡
20.440‡ 20.050 20.260* 20.495‡
MIS, Malnutrition Inflammation Score; TIBC, total ironbinding capacity; DW, dry weight. Pearson’s correlation method is used. Shown in each cell is the value of R. The MIS includes scores from 0 to 3, indicating normal to severely abnormal in each component. For detailed definitions of scores 0 to 3 in each component, please refer to the Appendix. *P , .05. †P , .01. ‡P , .001.
brief, there were also age differences in the correlation of dry weight with MIS components and relevant nutritional markers.
Multivariate Linear Regression of RIDWG and Dry Weight With MIS Components and Other Relevant Parameters To further differentiate the independent effect of RIDWG from that of dry weight, we reanalyzed the corresponding correlations in Tables 2 and 3 using multivariate linear regression, with RIDWG and dry weight regarded as independent variables. The results are presented in Tables 6 and 7. In the whole patient and the younger patient groups, the correlations between RIDWG and the MIS components were almost covered by those of dry weight. On the other hand, in the older patient group, the relationship was still significant between RIDWG and the severity of insufficient food intake, GI upset, functional
218
YANG ET AL
incapacity, and wasting of muscle and subcutaneous fat (Table 6). As shown in Table 7, RIDWG had its own significant correlations, independent of those of dry weight, with delivered dialysis dose and serum levels of phosphate and sodium in the whole, the younger, and the older patient groups.
Table 5. Correlations of DW With Relevant Parameters in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients Age ,65 Age $65 (n 5 164) (n 5 106) (n 5 58) Delivered dose of dialysis Daugirdas12 Kt/V Anthropometry MAMC Nutritional markers Albumin TIBC nPCR Laboratory data P Na Inflammation parameter C-reactive protein
20.639‡
0.201†
20.612‡ 20.670‡
0.159
Discussion
0.218
0.230† 0.382‡ 20.071
0.222* 0.018 0.395‡ 0.256 20.117 20.154
0.116 20.101
0.104 20.088
0.008 20.063
20.101
20.129
20.048
MAMC, derived mid-arm muscle circumference; TIBC, total iron-binding capacity; nPCR, normalized protein catabolic rate by volume; P, phosphate; DW, dry weight. Pearson’s correlation method is used. Shown in each cell is the value of R. *P , .05. †P , .01. ‡P , .001.
In the present study, we found that in younger HD patients, IDWG and RIDWG had almost no significant correlation with MIS components. In older patients, although IDWG also had almost no significant correlation with nutritional parameters, RIDWG had a positive correlation with severity of insufficient food intake, GI upset, functional incapacity, and wasting of muscle and subcutaneous fat (Table 2). Our findings suggest that in older HD patients, the greater the RIDWG, the poorer the nutritional status will be. On the other hand, we also found that dry weight was negatively correlated with severity of insufficient food intake and GI upset in older HD patients (Table 4). Because dry weight was the denominator and negatively correlated with RIDWG in older HD patients, the associations of higher RIDWG with higher MIS component
Table 6. Multivariate Linear Regression of RIDWG and DW With MIS Items in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients (n 5 164)
Change in dry weight Dietary intake Gastrointestinal symptoms Functional capacity Comorbidity Decrease or loss of fat Signs of muscle wasting Body mass index Serum albumin Serum TIBC Total score
Age ,65 (n 5 106)
Age $65 (n 5 58)
RIDWG
DW
RIDWG
DW
RIDWG
DW
20.145 0.135 0.188* 0.141 0.003 0.184† 0.092 0.061 0.011 20.035 0.111
20.158* 20.226† 20.169* 20.154* 20.255† 20.407‡ 20.399‡ 20.386‡ 20.227† 20.332‡ 20.466‡
20.137 0.068 0.145 0.135 0.022 0.110 0.065 20.024 0.056 20.103 0.055
20.139 20.041 0.046 20.072 20.237* 20.374‡ 20.309† 20.407‡ 20.242* 20.352‡ 20.443‡
20.106 0.303* 0.324* 0.307* 20.009 0.436‡ 0.283* 0.167 0.130 0.110 0.339†
20.108 20.302* 20.316* 20.072 20.243 20.298* 20.395† 20.384† 20.006 20.223 20.382†
RIDWG, relative interdialytic weight gain; MIS, Malnutrition Inflammation Score; TIBC, total iron-binding capacity; DW, dry weight. The RIDWG and DW are entered as independent variables, and other relevant parameters are entered as dependent variables. Shown in each cell is the value of the standardized coefficient (b). The MIS includes scores from 0 to 3, indicating normal to severely abnormal in each component. For detailed definitions of scores 0 to 3 in each component, please refer to the Appendix. *P , .05. †P , .01. ‡P , .001.
219
IDWG AND NUTRITION IN HD
Table 7. Multivariate Linear Regression of RIDWG and DW With Relevant Parameters in All Patients (n 5 164), Including Younger (Age ,65 Years) and Older (Age $65 Years) Patients All Patients (n 5 164)
Delivered dose of dialysis Daugirdas12 Kt/V Anthropometry MAMC Nutritional markers Albumin TIBC nPCR Laboratory data P Na Inflammation parameter C-reactive protein
Age ,65 (n 5 106)
Age $65 (n 5 58)
RIDWG
DW
RIDWG
DW
RIDWG
DW
20.237‡
20.683‡
20.206†
20.646‡
20.338†
20.784‡
0.173
0.082
0.246
0.112
0.220†
0.102
0.031 0.105 0.104
0.235† 0.399‡ 20.052
20.023 0.154 0.119
0.219* 0.417‡ 20.099
20.126 20.063 20.068
20.025 0.235 20.177
0.345‡ 20.323‡
0.174* 20.155*
0.275† 20.342‡
0.142 0.135
0.408† 20.280*
0.145 20.157
0.091
20.085
0.128
20.084
20.075
20.110
RIDWG, relative interdialytic weight gain; MAMC, derived mid-arm muscle circumference; TIBC, total iron-binding capacity; nPCR, normalized protein catabolic rate by volume; P, phosphate; DW, dry weight. The RIDWG and dry weight are entered as independent variables and other relevant parameters as dependent variables. Shown in each cell is the value of standardizied coefficient (b). *P , .05. †P , .01. ‡P , .001.
scores of food intake and GI upset would only be the reciprocal reflection of the associations between dry weight and MIS component scores. It is reasonable that older HD patients with insufficient food intake and more GI upset would have a lower dry weight and, consequently, higher RIDWG. However, as shown in Table 6, in the older patient group, the correlations of RIDWG were independent from those of dry weight, concerning severity of insufficient food intake, GI upset, functional incapacity, and wasting of muscle and subcutaneous fat. In regard to the discordances between IDWG and RIDWG in their correlations with MIS components, especially when age differences are considered, there are several possible explanations. First, younger patients would have ingested a more than adequate amount of food to maintain an acceptable nutritional status. Therefore, IDWG lost its correlation with other nutritional markers in younger HD patients. Second, as previously mentioned, the age differences of dry weight in the corresponding correlations with MIS components would be responsible to a certain extent. Third, the age difference in dietary intake could also be contributory. Regarding dietary intake status in the younger patient group (n 5 106), 93 patients had a score of 0, and 13 had a score of 1; in
the older patient group (n 5 58), 33 patients had a score of 0, 22 had a score of 1, and 3 had a score of 2. Thus the older patients had a higher rate of ingesting semifluid and full-fluid foods than did younger patients (c2 5 21.695, P , .001), compatible with the higher rate of dental problems in older patients. Fourth, malnourished older patients would have taken more liquid nutritional supplements,13 such as Nepro (Abbott Laboratories Services Corp., Columbus, OH). Fifth, malnourished older patients would have a higher rate of ingesting semifluid or full-fluid food via the nasogastric tube. Some studies found that IDWG was correlated negatively with patient age6,14,15 and positively with serum albumin levels6–8 in HD patients. Consequently, ignoring the confounding effect that younger patients actually have better nutritional status, IDWG suggested as a positive nutritional marker.16 Nevertheless, taking age differences into consideration is a logical first step, while investigating whether IDWG can be regarded as a nutritional marker warrants further study. In the present study, IDWG was correlated positively with serum albumin levels in the whole patient group (Table 3). However, the positive correlation did not exist in the younger or the older patient groups (Table 3). Therefore, in accordance with our previous
220
YANG ET AL
proposal, it is hard to present IDWG as a positive nutritional marker when patient age is considered. With regard to the relationship between IDWG and nPCR, a positive correlation was reported.8,14,15 However, there was no similar finding in the present study. Taking age difference into consideration, we found that serum albumin levels positively correlated with nPCR in the whole (r 5 0.289, P ,.001) and in the older (r 5 0.428, P 5 .001) patient groups, but not in the younger group (r 5 0.089, P 5.366). That is, the positive correlation between nPCR and albumin levels only existed in the older patient group. Therefore, we suggest that the results of previous studies concerning the relationships among nutritional markers be reexamined, taking age difference into consideration. Regardless of age difference, IDWG has a negative correlation with dialysis clearance (Daugirdas Kt/V), although RIDWG does not (Table 3). A similar finding was reported by Lopez-Gomez et al.6 If the confounding effects of dry weight shown in Table 5 are taken into account, the negative correlations of IDWGs would be offset by the negative correlations of dry weight. Therefore, it is not surprising that RIDWG has no significant correlation with dialysis clearance. Before the results of further study are made available, we assume that the negative correlation between IDWG and dialysis clearance can be partly, if not completely, explained by a lower average blood flow attributable to more episodes of interdialytic hypotension. In clinical practice, higher IDWG and RIDWG cause intradialytic hypotension more frequently.17,18 Logically, IDWG is assumed to have a positive correlation with the amount of food intake in HD patients,19,20 especially when patients become anuric after at least 1 year of HD therapy. The RIDWG is also assumed to correlate positively with the relative amounts of food intake in HD patients when individual dry weights are considered. The above inferences are indirectly supported by the findings of our study. As shown in Table 3, IDWG and RIDWG actually have a significantly positive correlation with serum phosphate levels (a positive indicator of food intake),21 and a negative correlation with serum sodium levels (a negative surrogate of sodium-free water intake). Lopez-Gomez JM et al reported similar findings.6 Furthermore, as seen in Table 1, the older patient group had significantly less IDWG and dry weight than the younger group. Meanwhile, there was
only a marginal difference in RIDWG (Table 1). Therefore, in accordance with previous studies,19,20 we agree that younger HD patients demonstrate more food intake than older patients. However, the relative amounts of food intake regarding respective dry weight would be similar in both groups of patients. In conclusion, we demonstrated the indispensable role of age difference when analyzing relationships among nutritional markers. Nevertheless, our findings here do not fully support the contention that IDWG reflects caloric and protein intake and can be regarded as a nutritional marker, as suggested in previous studies. Although IDWG and RIDWG could be positively correlated with the amount of food intake, we also found that RIDWG would be a negative nutritional maker, correlating with MIS in older HD patients. Perhaps high IDWG is associated with high salt and water intake rather than caloric and protein intake. This might especially be an issue in this study as this appears to be conducted in an Asian population. Chinese food is high in salt. Therefore, until more clinical evidence becomes available to differentiate food intake from nutritional gain, it remains controversial to regard IDWG and RIDWG as nutritional markers, especially when age difference is considered.
References 1. Sharp J, Wild MR, Gumley AI, Deighan CJ: A cognitive behavioral group approach to enhance adherence to hemodialysis fluid restrictions: A randomized controlled trial. Am J Kidney Dis 45:1046-1057, 2005 2. Sharp J, Wild MR, Gumley AI: A systematic review of psychological interventions for the treatment of nonadherence to fluid-intake restrictions in people receiving hemodialysis. Am J Kidney Dis 45:15-27, 2005 3. Foley RN, Herzog CA, Collins AJ: United States Renal Data System. Blood pressure and long-term mortality in United States hemodialysis patients: USRDS Waves 3 and 4 Study. Kidney Int 62:1784-1790, 2002 4. Hecking E, Bragg-Gresham JL, Rayner HC, et al: Haemodialysis prescription, adherence and nutritional indicators in five European countries: Results from the Dialysis Outcomes and Practice Patterns Study (DOPPS). Nephrol Dial Transplant 19: 100-107, 2004 5. Rahman M, Fu P, Sehgal AR, Smith MC: Interdialytic weight gain, compliance with dialysis regimen, and age are independent predictors of blood pressure in hemodialysis patients. Am J Kidney Dis 35:257-265, 2000 6. Lopez-Gomez JM, Villaverde M, Jofre R, RodriguezBenitez P, Perez-Garcia R: Interdialytic weight gain as a marker of blood pressure, nutrition, and survival in hemodialysis patients. Kidney Int 93(suppl):S63-S68, 2005
IDWG AND NUTRITION IN HD 7. Szczech LA, Reddan DN, Klassen PS, et al: Interactions between dialysis-related volume exposures, nutritional surrogates and mortality among ESRD patients. Nephrol Dial Transplant 18: 1585-1591, 2003 8. Testa A, Beaud JM: The other side of the coin: interdialytic weight gain as an index of good nutrition. Am J Kidney Dis 31: 830-834, 1998 9. Enia G, Sicuso C, Alati G, et al: Subjective global assessment of nutrition in dialysis patients. Nephrol Dial Transplant 8: 1094-2008, 1993 10. Kalantar-Zadeh K, Kopple JD, Block G, et al: A malnutrition-inflammation score is correlated with morbidity and mortality in maintenance hemodialysis patients. Am J Kidney Dis 38: 1251-1263, 2001 11. Morduchowicz G, Winkler J, Zabludowski JR, Boner G: Effects of residual renal function in haemodialysis patients. Int Urol Nephrol 26:125-131, 1994 12. Daugirdas JT: Second generation logarithmic estimates of single-pool variable volume Kt/V: An analysis of error. J Am Soc Nephrol 4:1205-1213, 1993 13. Caglar K, Fedje L, Dimmitt R, et al: Therapeutic effects of oral nutritional supplementation during hemodialysis. Kidney Int 62:1054-1059, 2002 14. Kimmel PL, Varela MP, Peterson RA, et al: Interdialytic weight gain and survival in hemodialysis patients: Effects of duration of ESRD and diabetes mellitus. Kidney Int 57:1141-1151, 2000 15. Ifudu O, Uribarri J, Rajwani I, et al: Relation between interdialytic weight gain, body weight and nutrition in hemodialysis patients. Am J Nephrol 22:363-368, 2002 16. Sarkar SR, Kotanko P, Levin NW: Interdialytic weight gain: Implications in hemodialysis patients. Semin Dial 19:429-433, 2006 17. Takeda A, Toda T, Fujii T, Sasaki S, Matsui N: Can predialysis hypertension prevent intradialytic hypotension in hemodialysis patients? Nephron 103:c137-c143, 2006 18. Kursat S, Ozgur B, Alici T: Effect of ultrafiltration on blood pressure variability in hemodialysis patients. Clin Nephrol 59: 289-292, 2003 19. Testa A, Plou A: Clinical determinants of interdialytic weight gain. J Ren Nutr 11:155-160, 2001 20. Bossola M, Muscaritoli M, Tazza L, et al: Variables associated with reduced dietary intake in hemodialysis patients. J Ren Nutr 15:244-252, 2005 21. Levy J, Morgan J, Brown E: Assessment of nutitional status: biochemical. In: Levy J, Morgan J, Brown E (eds.): Oxford Handbook of Dialysis. New York: Oxford University Press, Inc., 2001, pp 388-389
Appendix
The following are details of the definitions of each component of the Malnutrition Inflammation Score copied from the original proposal of Kalantar-Zadeh et al.10 Weight change is determined as the change in edema-free postdialysis body weight in the previous 6 months. The lowest score (0) is given if weight loss is ,0.5 kg, or there is an increase in body weight. A score of 1 indicates a minor loss of at least 0.5 kg, but ,1.0 kg. A score of 2 is given for weight loss of at least 1.0 kg, but ,5% of body weight, and a score of 3 indicates a weight
221
loss of $5%. Dietary intake is scored as 0 if it involves the usual intake of solid foods, with no recent decrease in the amount or quality of meals. A score of 1 indicates a slightly suboptimal solid diet, a score of 2 indicates a full-liquid diet or moderate decrease in food intake, and a score of 3 indicates a daily nutrient intake that would be incompatible with life on a long-term basis. Gastrointestinal symptoms are scored as 0 if the patient has a good appetite and no GI symptoms. A score of 1 indicates mildly decreased appetite or mild nausea. A score of 2 indicates occasional vomiting or other moderate GI symptoms, such as abdominal pain. A score of 3 indicates diarrhea, frequent vomiting, or severe anorexia. Functional capacity is scored as 0 for normal functional capacity or a considerable improvement in the level of previous functional impairment. A score of 1 indicates mild or occasional difficulty with baseline ambulation or feeling tired frequently. A score of 2 indicates difficulty with independent activities. A score of 3 indicates restriction to light activity or a persistent bedbound or chairbound state. Comorbidity is scored as 0 if there are no other medical illnesses and the patient has undergone hemodialysis therapy for ,1 year; as 1 for mild comorbidity, excluding such major comorbid conditions (MCCs) as congestive heart failure class III or IV, severe coronary artery disease, clinically evident acquired immunodeficiency syndrome, moderate to severe chronic obstructive pulmonary disease, and metastatic malignancies, or dialysis therapy for 1 to 4 years; as 2 for moderate comorbidity (including one of the diseases listed under MCCs) or dialysis therapy for .4 years; and as 3 for $2 MCCs. The existence of diabetes per se is not accounted for if the previously mentioned comorbidities do not exist. Instead, comorbidities that may pose a risk for poor outcomes in patients with diabetes are examined individually. Body-fat stores are scored by assessing subcutaneous fat deposition in four body areas: below the eyes, the triceps, the biceps, and the chest. Signs of muscle wasting are determined by briefly examining seven sites: the temple, clavicle, scapula, ribs (intercostal spaces), quadriceps, knee, and interosseous muscles. For each of these two components, a score of 0 to 3, representing normal to severe changes, is assigned according to conventional SGA guidelines. Body mass index was graded at four levels (0 to 3), representing BMI .20, 18 to 19.99, 16 to 17.99, and ,16 kg/m2, respectively. A score of 0
222
YANG ET AL
in serum albumin means that the level is $40 g/L (or $4.0 g/dL). A score of 1 indicates 35 to 39 g/L (or 3.5 to 3.9 g/dL). A score of 2 indicates 30 to 34 g/L (or 3.0 to 3.4 g/dL). A score of score 3 indicates ,30 g/L (or ,3.0 g/dL).
The scores 0 to 3 in TIBC represent TIBC $2.50 g/L (or 250 mg/dL), 2.00 to 2.49 g/L (or 200 to 249 mg/dL), 1.50 to1.99 g/L (or 150 to 199 mg/dL), and ,1.50 g/L (or ,150 mg/dL), respectively.